Robbe Roxy 2834/10 rewinding

[ZuvieleTeile]

HELP!!!!!

Ralph, I need your help, please!!

Got a tiny little thing in the mail. Looks like a motor.
It's called Robbe-Roxy 2834/10.
Those are not the stator numbers, but the outside measurements instead . . .

It is a 12N14P motor with a stator of 22mm dia and 16mm height.
Original Kv is 880.
This little boy looks a little bit burned.
Got it easily apart with almost just looking at it. . .
Went through your site with the 12N14P windings. Lots of info there, but maybe you can chime in here and let me know, what the best re-winding scheme for this tiny baby would be.
Not sure what kind of wire I need. My smallest one is 0.8mm though.

It might not be worth of re-winding it, but hey, it is a challenge. . .

[ZuvieleTeile]

Nobody chimes in??
C'mon guys, anybody?

Since it is a 12N14P motor, I can use the beloved AabBCc/aABbcC scheme, but I am not sure, if it results in the lowest number of turns.
It will run on a Robbe ESC, so I suppose it does not matter, if I wind it in D or Y.
Any opinion will be welcome . . .

[Vinger]

Wound a few motors for indoor foamies and used this pattern. .8mm might be too thick, I normally use .4-.6mm wire. About 12 turns for this size will get you 1000KV

[ZuvieleTeile]

Thanks Vinger,
I am just contemplating which termination gives me the least amount of turns for a given Kv.
Have to dig into powerditto again.
I think it is D.

[Vinger]

No, Delta will need more turns for same KV.

Great info on smaller motors here as well. http://www.bavaria-direct.co.za/models/motor_info.htm

[ZuvieleTeile]

Originally Posted by Vinger

No, Delta will need more turns for same KV.

It depends . . .
If you take a pure D termination and a pure Y termination, yes.
Y runs 1.73 times slower than D.
But . . .
Since we are doing mostly YY, we need about 15% more turns in YY to slow the Kv down to a D termination.

[Stolla]

correct one of the reasons I only do yy, to get more copper on for same kv, wonder though what real benefits are on these small motors, imagine rewinding my hexa's 6 motors!

[Vinger]

You guys are confusing me now! (very easy to do!)

As I understand it YY needs 1.73 turns LESS than DELTA for the same KV or nSpec. Therefore as ZT has asked to get the LEAST amount of turns for the KV, that means that a YY will be the termination for what he want?

Maybe my understanding of German and Google translate have taken me in the wrong direction!!!

[ZuvieleTeile]

@ Stolla,
this little MK3638 motor is also a 14 pole motor and probably a 12 tooth stator.
It'll be hard to get more power out of those imho because they are already absolutely filled with wires/copper.
As with the little Roxy, I am doing right now, the numbers of the MK are probably the outside measurements and not the stator dimensions.
I am doing the rewind only because of the damage and out of curiosity.

@ Vinger
If you take a D terminated motor and compare it to a Y terminated motor, then the Y would run much slower, or you use less turns to achieve the same Kv.
In my post # 14 I was thinking of a YY termination.
Let's make an example here (I hope, my understanding of the issue is correct)

Let's take a 12N10P motor, which was terminated in D, like most of them are with the factory winding. And let's say, it has a Kv of 500.
We count the D windings and come up with 12 turns per group. (6+6)
If we would rewind that very motor and do it in Y with the same number of turns, the motor would have a Kv of 500/1.73 = 289
In order to get back to our 500Kv, we would have to reduce the number of turns of the Y termination by the factor 1.73. That would get us to 7 turns (3+4)
Or we divide the stator into 2 motors, which cuts the turns between the phases in half and doubles the Kv again. That would leave us with a Kv of 578.
In order to get it down to our target of 500Kv, we have to do around 15% more turns.

The whole thing can also simply be calculated:
D divided by 1.73 gives us the number of turns for a Y termination and then times 2, the number of the turns for the YY termination.
This gives us 15% more Kv or we have to increase the turns by 15 % to get down to the same Kv again

Relationship Y to D = 1/1.73
Relationship YY to D = 2/1.73

That's at least how I understand the whole confusing thing . . .
The nice thing about this forum is, that we are able to exchange our thoughts and theories here and somebody might jump in and have a better or different idea.

If I explained it correct and some people got something out of this, that gives me a good feeling.
If I did it wrong or there are some kinks in my explanantion, then Ralph or some other knowledgable person might jump in and correct me. . . . and I learn something.
In any case, it is a win-win situation.

[HeliFX]

ZT,

Holy crap that motor had a lot of epoxy

[ZuvieleTeile]

Taking off the wires was quite an undertaking. They used lots of epoxy.

But finally the little beast had to give up and the naked stator was in front of me.
Needed some minor repairs of the coating with electrical varnish, but was good to be wound for a test.
Found some thin 0.4 mm wire at Radio Shack and got my stuff together.
Wanted to try a D winding and dug out Ralph's scheme for the optimal and most effective winding.
I went for 5 turns per tooth just to get some feel, how this little thing is behaving.
Stator height is 16 mm ( btw. they used 45 plates, which is about 0.35 mm per plate)

To calculate the required wire per phase I did the following calculation.
16 x 2 plus 6 for head and bottom = 35mm or 3.5 cm.
I had 4 teeth per group to wind. 3.5 x 4 = 14 cm
I went for 5 turns per tooth. 14 x 5 = 70 cm
For the beginnings and ends I thought 7 cm would be enough. 70 + 14 = 84 cm
And then I added about 6 cm for the transitions between the groups. 84 + 6 = 90 cm

I cut 3 times the 90 cm and started winding away.

I wound exactly according the scheme and discovered, that Ralph put a lot of thoughts into this.
There are only crossings of two wires at one time. You have to be very thorough and really look at how the turns are going and don't confuse the cw and ccw turns/groups.
First it makes no sense, but after carfully examining the groups and taking into account where the power comes from and is going to, then slowly you'll see a glimpse of light at the end of the tunnel.

The other nice feature of that scheme is, that each wire-pair, which creates the ESC connection, comes out of one slot. So there is no confusion, which wires have to be soldered together, like it happened to me on my HK4015 project.

Put it all together, mounted it on my little test stand, connected a 3 cell Lipo and and gave it power.
Started nicely and run 21400 RpM on 11.3 Volt
That's 1880 Kv.
Huuhh, way too high. The original was 880 Kv and the owner would like to have it that way back.
In order to do that I would have to more than double the turns.
Difficult task.
That's where I am right now and still scratching my head.
.

[HeliFX]

ZT,

thought I point out your error in case you fall short on the wire.
I'm sure it won't matter much

16 x 2 plus 6 for head and bottom = 35mm ( should be 38 mm)

Originally Posted by ZuvieleTeile

Taking off the wires was quite an undertaking. They used lots of epoxy.

But finally the little beast had to give up and the naked stator was in front of me.
Needed some minor repairs of the coating with electrical varnish, but was good to be wound for a test.
Found some thin 0.4 mm wire at Radio Shack and got my stuff together.
Wanted to try a D winding and dug out Ralph's scheme for the optimal and most effective winding.
I went for 5 turns per tooth just to get some feel, how this little thing is behaving.
Stator height is 16 mm ( btw. they used 45 plates, which is about 0.35 mm per plate)

To calculate the required wire per phase I did the following calculation.
16 x 2 plus 6 for head and bottom = 35mm or 3.5 cm.
I had 4 teeth per group to wind. 3.5 x 4 = 14 cm
I went for 5 turns per tooth. 14 x 5 = 70 cm
For the beginnings and ends I thought 7 cm would be enough. 70 + 14 = 84 cm
And then I added about 6 cm for the transitions between the groups. 84 + 6 = 90 cm

I cut 3 times the 90 cm and started winding away.

I wound exactly according the scheme and discovered, that Ralph put a lot of thoughts into this.
There are only crossings of two wires at one time. You have to be very thorough and really look at how the turns are going and don't confuse the cw and ccw turns/groups.
First it makes no sense, but after carfully examining the groups and taking into account where the power comes from and is going to, then slowly you'll see a glimpse of light at the end of the tunnel.

The other nice feature of that scheme is, that each wire-pair, which creates the ESC connection, comes out of one slot. So there is no confusion, which wires have to be soldered together, like it happened to me on my HK4015 project.

Put it all together, mounted it on my little test stand, connected a 3 cell Lipo and and gave it power.
Started nicely and run 21400 RpM on 11.3 Volt
That's 1880 Kv.
Huuhh, way too high. The original was 880 Kv and the owner would like to have it that way back.
In order to do that I would have to more than double the turns.
Difficult task.
That's where I am right now and still scratching my head.
.

[ZuvieleTeile]

Thanks HeliFX,
was just a typo.
In my original, I did it right.
Next time I would give it some more anyway, because it was a little bit tight at the end.
I didn't do a thorough wind and wasted some wire on the way probably.
It was just for the Kv determination.

[HeliFX]

I'll still grade a A- for math

I also add another 2 to 5% on the overall length to allow for copper thickness and any transitions.

Originally Posted by ZuvieleTeile

Thanks HeliFX,
was just a typo.
In my original, I did it right.
Next time I would give it some more anyway, because it was a little bit tight at the end.
I didn't do a thorough wind and wasted some wire on the way probably.
It was just for the Kv determination.

[Vinger]

Thanks ZT, your explanation helped my confused brain on the correct path!!!

[ZuvieleTeile]

@ Ralph,
I am still in the process figuring out how I can wind this little motor with less turns, in order to get down to the 800-900 Kv range.
D and YY proved itself not to be feasible. I would be needing to wind about 10 to 11 turns per tooth and the wire for the available little space had to be very thin. Probably 0.4 mm.

How about I wind it in a plain Y and not YY. That would double the turns per phase, get the KV down and I could use thicker wire?
What would be the disadvantage here?
I think I got the proper scheme out of your site: Distributed LRK.

I am grateful for any advise . . .

[ZuvieleTeile]

Little addendum to my # 16,
playing around with the winding calculator and looking at the different pictures I just realized that I can take my little YY half stator winding and unsolder the starpoint, solder some additional wire at the ends and keep winding the other three groups in order to get to the Y winding.
The new ends would be the starpoint then.
This would double the number of windings like stated above and get my Kv down.
Still not sure, if it has any disadvantages though . . .
I consider myself still rookie in this regard.
Any oppinion Ralph???

[ZuvieleTeile]

The soldering of additional wires to the old star-point ends turned out to be very messy and not a good choice.
So, I cut all wires and started from scratch a nice Y test winding.
Still with he ugly green wire, which is hard to see, if you want to straighten something out between the slots. . .
Kv came out nicely at around 900, which ws the target.

Now, the million Dollar question is:
Which termination gives me the least resistance and the best copper filling?
The YY with thinner wires, or the Y with a thicker wire?
Everything else being equal, the resistance and copper amount should be similar, but there must be an advantage from one over the other.
I could do a gazillion winds with different wire ranging from 20 to 26 AWG, but first of all I don't have those wires and secondly it will be too much effort.
I am still counting on some experienced winders here to chime in and let me participate in their knowledge of the subject. . .

[ZuvieleTeile]

Continuing my monologue here,
After some more reading, head scratching and changing the batteries in my calculator I come up with the following:
I have to see what kind of space I have available in the slots and how many turns give me my target Kv.
Considering my stock of available wires, I have two choices now.
Either wind it in Y 6+6 or in D 10+10.
Both would be close to my target Kv.
I brought some pieces of my wire to a glow and scraped off the insulation carefully in order to get some real copper numbers.
The thinner wire has a dia of 0.41 mm with iso and 0.38 without iso.
The thicker wire has a dia of 0.67 with iso and 0.63 without.

A quick calculation with my freshly powered calculator revealed the following:
The thinner wire has 0.113 sq./mm
The thicker one has 0.312 sq./mm

That would leave me for the D winding with: 10 x 0.113 = 1.13 sq./mm total copper/tooth.
And for the Y winding: 6 x 0.312 = 1.87 total copper/tooth.

That looks like a win for the Y winding. (Except our math professor from Canada finds some errors again)

The 6 turns don't fit completely into the slot, so I have to wind # 6 on top of 4 and 5.
Then jump to the inside of the slot and start the same thing again.
If I get this one managed, I can proceed with the same wire on the opposite side of the stator with the next group for this phase.
We'll see